本文已被:浏览 466次 下载 3次
投稿时间:2019-09-29 修订日期:2019-10-30
投稿时间:2019-09-29 修订日期:2019-10-30
中文摘要: 传统的塔筒倾覆在线监测系统往往存在数据采集、分析单一、非专业人员很难看懂和理解、运维管理模式落后等缺陷,为了解决这些问题,本文通过研究人工智能技术,综合塔筒倾斜、基础水平、风速、风向、功率等关联信息,运用机器学习算法构建塔筒倾覆故障模型,计算反映设备状态的模型特征值,检测模型特征值的变化趋势,实现塔筒倾覆状态的在线检测和劣化过程的早期预警。同时本文还重点介绍了基于人工智能的风机塔筒倾覆智能预警系统建设目标、系统架构、诊断流程和系统开发平台,通过此系统不断提高对数据挖掘技术的能力,对风机健康状况进行更为精准的判定及故障隐患的及时发现,保障风机基础和塔筒的安全运行,防止风机发生倾斜或倒塌等重大安全生产事故。
中文关键词: 人工智能,塔筒倾覆,特征值,预警
Abstract:The traditional on-line monitoring system for tower capsizing often has some shortcomings, such as data acquisition, single analysis, difficulty for non-professionals to understand and understand, and backward operation and maintenance management mode. In order to solve these problems, this paper synthesizes the related information of tower inclination, basic level, wind speed, wind direction and power by studying artificial intelligence technology. The machine learning algorithm is used to construct the tower overturning fault model, calculate the eigenvalues of the model reflecting the state of the equipment, detect the changing trend of the eigenvalues of the model, and realize on-line detection of the tower overturning state and early warning of the deterioration process. At the same time, this paper also focuses on the AI-based wind turbine tower overturning intelligent early warning system construction objectives, system architecture, diagnosis process and system development platform. Through this system, the ability of data mining technology is continuously improved, and the health status of the wind turbine is more accurate judgement and timely detection of hidden troubles, Safe operation of fan foundation and tower barrel is ensured, and major safety accidents such as inclination or collapse of fan are prevented.
文章编号: 中图分类号: 文献标志码:
基金项目:
作者 | 单位 | |
陈万勋* | 中国电建集团贵州工程有限公司 贵州省贵阳市 | lcb08@163.com |
刘春波 | 北京奥技异电气技术研究所有限公司 北京市 | |
赵坚强 | 北京奥技异电气技术研究所有限公司 北京市 |
引用文本: